Sleep

Evaluating Electroencephalogram-Based Predictive Model for Drowsiness Measurement to Reduce Accident Risk in Active Individuals: Protocol for a Preliminary Monocentric Study.

TL;DR

This protocol describes a single-center, nonrandomized, single-group study designed to validate continuous or predictive methods for assessing drowsiness using automated analysis of a limited number of EEG channels in 40 healthy volunteers exposed to two sleep deprivation conditions simulating real-world occupational scenarios.

Key Findings

The study is designed as a single-center, nonrandomized, single-group investigation enrolling 40 healthy volunteers.

  • Participants are healthy volunteers, described as 'active individuals' exposed to drowsiness-inducing conditions.
  • The study is monocentric (single-center) and does not include randomization or a control group.
  • Recruitment began in March 2023 and was completed in May 2025.
  • A database lock was set for June 2025, with data analysis starting in June 2025 and described as 'still ongoing' at time of publication.

Participants are exposed to two sleep deprivation conditions designed to simulate real-world occupational scenarios.

  • The two conditions are intended to reflect voluntary behaviors and socioeconomic factors such as social jetlag and shift work.
  • Insufficient or disrupted sleep resulting from these conditions is the mechanism by which drowsiness is induced.
  • The conditions are meant to reflect contexts where sustained vigilance is critical, including occupational and driving environments.

The primary outcome measure is the Objective Sleepiness Scale (OSS) and its automated analysis, with a focus on its ability to measure objective wakefulness as assessed by the maintenance of wakefulness test (MWT).

  • The OSS is evaluated through automated analysis of a limited number of EEG channels.
  • The MWT serves as the reference standard for objective wakefulness measurement.
  • The study aims to validate continuous or predictive methods for drowsiness assessment using the OSS.

Secondary outcomes include a broad multimodal battery covering EEG markers, subjective and objective sleepiness, simulated driving performance, cognitive assessments, sleep quality and quantity, and mind-wandering.

  • Multimodal resting-state EEG markers are included as secondary outcomes.
  • Cognitive domains assessed include attention, executive function, and vigilance.
  • A simulated driving task is included to evaluate real-world safety-relevant performance.
  • Sleep quality, sleep quantity, and mind-wandering are also systematically measured.
  • The influence of sociodemographic and clinical variables on the measurement and prediction of drowsiness will be systematically examined.

The study received funding from Physip and ANR (Agence Nationale de la Recherche) in 2019, with ethical approval granted in May 2022.

  • Funding sources include the private company Physip and the French National Research Agency (ANR).
  • Ethical committee approval was granted by the Comité de Protection des Personnes (Committee for the Protection of Persons) in May 2022.
  • The gap between funding (2019) and ethical approval (2022) represents approximately three years.

The study aims to lay the groundwork for proactive drowsiness management strategies in occupational, transportation, and clinical settings by validating novel EEG-based measures.

  • The target applications span three domains: occupational, transportation, and clinical settings.
  • The approach focuses on automated analysis of a limited number of EEG channels, emphasizing practical deployability.
  • The study frames drowsiness as a serious safety risk that impairs alertness, slows reaction times, and increases accident likelihood.
  • Developing 'automatic and easy-to-implement tools for drowsiness detection or prediction' is stated as essential for managing sleepy patients or high-risk environments.

What This Means

This research describes the protocol for a study investigating whether brain wave (EEG) measurements can automatically and accurately detect or predict drowsiness in people who are sleep-deprived. The study enrolled 40 healthy adults and exposed them to two different types of sleep loss meant to mirror common real-world situations, such as working night shifts or staying up late due to social obligations. Researchers are testing whether a tool called the Objective Sleepiness Scale (OSS), which analyzes EEG signals automatically, can reliably measure how awake or sleepy someone is, compared to a standard clinical test called the Maintenance of Wakefulness Test. The study also collects a wide range of additional data, including brain activity at rest, performance on a driving simulator, attention and thinking ability tests, and self-reported sleepiness, to build a comprehensive picture of how drowsiness affects functioning. The goal is to see whether a small number of EEG sensors — making the technology more practical to use outside of a laboratory — can still provide accurate, real-time drowsiness information. This research suggests that if EEG-based drowsiness detection can be validated with a simplified setup, it could eventually be used in workplaces, vehicles, or clinics to identify dangerously sleepy individuals before accidents occur. This would be particularly relevant for professions requiring sustained attention, such as drivers, pilots, or medical workers, and could also help clinicians monitor patients with sleep disorders. The data analysis was still in progress at the time the protocol paper was published, so results have not yet been reported.

Have a question about this study?

Citation

Boitard C, Mazurie Z, Sadatnejad K, Coelho J, Sagaspe P, Lenoir J, et al.. (2026). Evaluating Electroencephalogram-Based Predictive Model for Drowsiness Measurement to Reduce Accident Risk in Active Individuals: Protocol for a Preliminary Monocentric Study.. JMIR research protocols. https://doi.org/10.2196/83969